Key result
A classification approach using hybrid features and a Random Forest classifier achieved an accuracy of 98.29% in detecting ECG arrhythmias from the MIT-BIH database.
Why the study?
Automatic detection of ECG signal characteristics is important for efficiently identifying heart diseases such as arrhythmia or ischemia.
Absolute Event Rate: 98.29% vs 94.29%
A machine learning approach using hybrid feature extraction and a random forest classifier achieved 98.29% accuracy in classifying ECG arrhythmias.
May aid arrhythmia detection via ECG transforms; leaves open prospective validation before clinical use.
In recent years, several methods have been established for the automatic detection of the characteristics of electrocardiogram (ECG) signals based on mathematical tools, among which the Fourier Transform, the Continuous Wavelet Transform (CWT), the Discrete Wavelet Transform (DWT) stand out among others. This type of procedure is important because it can be more efficiently detected if a certain patient has heart disease, such as an arrhythmia or ischemia. The primary goal of this paper is the development and implementation of a classification approach based on hybrid features i. e., Dual-Tree Complex Wavelet Transform (DTCWT), SVD-Entropy, Autoregressive modeling, and Multifractal analysis based feature extraction. To accomplish the better classification, the extracted features are further classified by Random Forest Classifier, K-Nearest Neighbors (KNN), and Bayesian Optimized-KNN classifiers utilizing the MIT-BIH database. Highest accuracy achieved in random forest classifier is 98.29 %.The results obtained show the feasibility and practical efficiency of the methodology as a tool to aid in the diagnosis of heart disease in hospital environments.
No takes yet. Share an insight, caveat, or question.
A 2021 study studied Cardiac arrhythmias (n=48). Hybrid feature extraction with Random Forest Classifier vs. K-Nearest Neighbors (KNN) and Bayesian Optimized-KNN classifiers was evaluated on Classification accuracy. A classification approach using hybrid features and a Random Forest classifier achieved an accuracy of 98.29% in detecting ECG arrhythmias from the MIT-BIH database.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: